tensorflow深度学习之生成数据集（二）

import os

import numpy as np

import tensorflow as tf

import input_data

import model

N_CLASSES = 2  #数据集分两类

IMG_W = 64  # 图片的高

IMG_H = 64 # 图片的宽

BATCH_SIZE = 16

CAPACITY = 1000

MAX_STEP = 10000 # 学习的步长

learning_rate = 0.0001 # 学习率

def run_training():

    # you need to change the directories to yours.

    train_dir = '/Users/Desktop/cd/cd/Far_1/'  #主要说下这个文件夹里边的图片 分成两类 一类是带image的图片名称， 一类是不带。。  图片的名称叫什么都行，学习特征两类，多类，都可以，需要自行修改代码。我是参考识别猫和狗的代码。。

    logs_train_dir = '/Users/Desktop/cd/cd/logs' #生成的日志文件，数据集和tensorflow学习的效率，可以使用 tensorbord进行查看

    train, train_label = input_data.get_files(train_dir)

    train_batch, train_label_batch = input_data.get_batch(train,

                                                          train_label,

                                                          IMG_W,

                                                          IMG_H,

                                                          BATCH_SIZE,

                                                          CAPACITY)     

    train_logits = model.inference(train_batch, BATCH_SIZE, N_CLASSES)

    train_loss = model.losses(train_logits, train_label_batch)       

    train_op = model.trainning(train_loss, learning_rate)

    train__acc = model.evaluation(train_logits, train_label_batch)

    summary_op = tf.summary.merge_all()

    sess = tf.Session()

    train_writer = tf.summary.FileWriter(logs_train_dir, sess.graph)

    saver = tf.train.Saver()

    sess.run(tf.global_variables_initializer())

    coord = tf.train.Coordinator()

    threads = tf.train.start_queue_runners(sess=sess, coord=coord)

    try:

        for step in np.arange(MAX_STEP):

            if coord.should_stop():

                    break

            _, tra_loss, tra_acc = sess.run([train_op, train_loss, train__acc])

            if step % 50 == 0:

                print('Step %d, train loss = %.2f, train accuracy = %.2f%%' %(step, tra_loss, tra_acc*100.0))

                summary_str = sess.run(summary_op)

                train_writer.add_summary(summary_str, step)

            if step % 2000 == 0 or (step + 1) == MAX_STEP:

                checkpoint_path = os.path.join(logs_train_dir, 'model.ckpt')

                saver.save(sess, checkpoint_path, global_step=step)

    except tf.errors.OutOfRangeError:

        print('Done training -- epoch limit reached')

    finally:

        coord.request_stop()

    coord.join(threads)

    sess.close()